Environment-friendly packaging box rapid forming method and system based on biodegradable material

Abstract

The invention relates to the technical field of 3D printing, and discloses an environment-friendly packaging box rapid forming method and system based on biodegradable materials, and the method comprises the steps that a thermal-structure coupling finite element model is established; based on the thermal-structural coupling model, a multi-objective optimization model is constructed, and structural design parameters and initial printing path configuration are obtained; training a printing control strategy by utilizing a strategy optimization algorithm; according to the structural design parameters, the printing path configuration and the printing control strategy, a layer-by-layer printing path and a supporting structure configuration scheme matched with the printing object are generated; and the 3D printing process is executed, and closed-loop adjustment is conducted according to the control strategy. According to the method, the thermal-structural coupling finite element model based on the biodegradable material is constructed, and changes of a temperature field, a stress field and a thermal strain field in the printing process are accurately simulated, so that a printing path and supporting structure configuration are optimized, the printing precision is improved, structural instability caused by thermal deformation and interlayer stress is avoided, and the strength and stability of an object are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of 3D printing, and specifically to a rapid prototyping method and system for an environmentally friendly packaging box based on biodegradable materials. Background Art

[0002] With the continuous improvement of environmental awareness, biodegradable materials have gradually become the focus of attention in the field of environmental protection packaging. Against this background, 3D printing technology has been widely used in the design and manufacture of environmentally friendly packaging boxes due to its high efficiency, flexibility, and customization characteristics. As a new type of green consumable for 3D printing, biodegradable materials can decompose under natural conditions, reduce plastic pollution, and meet the requirements of sustainable development.

[0003] Firstly, the thermophysical properties of biodegradable materials in 3D printing are relatively complex. Compared with traditional plastic materials, parameters such as thermal conductivity, specific heat capacity, and elastic modulus of biodegradable materials show obvious non-linearity during the printing process with the change of temperature. This characteristic makes it difficult to predict the distribution of temperature field and stress field during the printing process, and thus easily leads to problems such as thermal deformation and warping of the printed object, seriously affecting the quality and stability of the environmentally friendly packaging box.

[0004] Secondly, traditional methods often rely on simple slicing algorithms or empirical designs, and do not fully consider the thermal stress distribution and heat conduction characteristics of biodegradable materials. When printing environmentally friendly packaging boxes with complex structures or overhanging parts, existing technologies are difficult to effectively generate support structures that meet the mechanical property requirements, resulting in insufficient or unstable support during the printing process, and thus affecting the final printing quality. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technology, the present invention provides a rapid prototyping method and system for an environmentally friendly packaging box based on biodegradable materials, which solves the problems in the existing technology, such as the complex thermophysical properties of materials, thermal deformation and warping problems during the printing process, and the inability of existing path planning and support structure design to effectively handle the printing of complex structures.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A rapid prototyping method for an environmentally friendly packaging box based on biodegradable materials, comprising the following steps:

[0007] Collect the thermal conductivity, specific heat capacity, elastic modulus, and density parameters of the selected biodegradable material, and establish a thermal-structural coupling finite element model based on this material;

[0008] Based on the thermal-structural coupling model, construct a multi-objective optimization model with printing time, thermal field uniformity, forming strength, and material consumption as objectives, optimize the structural parameters of the packaging structure to be printed, and obtain the structural design parameters and the initial printing path configuration;

[0009] Based on the structural design parameters, a reinforcement learning environment is constructed, with the state variables during the printing process as inputs and the control variables as actions. A printing control strategy is trained using a policy optimization algorithm to achieve dynamic regulation of the printing process.

[0010] According to the structural design parameters, printing path configuration, and printing control strategy, a layer-by-layer printing path and support structure configuration scheme matching the printed object are generated.

[0011] Execute the 3D printing process according to the layer-by-layer printing path and support structure configuration scheme. During the process, print status data is collected in real time, and closed-loop adjustment is performed based on the control strategy.

[0012] Preferably, the thermo-structural coupling finite element model considers the thermal conductivity, specific heat capacity, and elastic modulus of the biodegradable material that vary with temperature during the printing process, and outputs the temperature field, stress field, and thermal strain field.

[0013] Preferably, the multi-objective optimization model is solved using a multi-objective optimization algorithm, and the optimization objectives include:

[0014] Minimization of printing time;

[0015] Minimization of the temperature difference between layers;

[0016] Maximization of structural stiffness;

[0017] Minimization of material usage.

[0018] Preferably, the design variables of the optimization model include printing layer thickness, filling density, local wall thickness variation, and ventilation structure layout parameters, and the packaging structure parameters include printing thickness, support structure layout, wall thickness, and ventilation structure configuration.

[0019] Preferably, the state variables include the real-time temperature of the printing nozzle, material extrusion rate, printing speed, hot bed temperature, and printing layer height; the control variables include nozzle heating power, air cooling speed, print head speed, and feeding rate.

[0020] Preferably, the reinforcement learning control strategy is trained using the Proximal Policy Optimization algorithm, and policy convergence is achieved through iterative simulation of the printing process.

[0021] Preferably, the printing path is generated based on the results of structural topology optimization and thermal simulation feedback, and the support structure is preferentially arranged in the thermally deformed sensitive area and the overhanging structure area.

[0022] Preferably, the closed-loop adjustment is performed by collecting the printing temperature, pressure, extrusion flow rate, and interlayer misalignment in real time, and adjusting the printing parameters based on the printing control strategy to reduce the risk of interlayer cracking and warping.

[0023] Preferably, the collected printing status data includes a thermocouple sensor, an infrared thermometer, and a displacement sensor.

[0024] An environmentally friendly packaging box rapid prototyping system based on biodegradable materials, comprising:

[0025] A material modeling module that establishes a thermal-structural coupling finite element model based on the thermal conductivity, specific heat capacity, elastic modulus, and density parameters of biodegradable materials;

[0026] A structural optimization module that constructs a multi-objective optimization model based on the thermal-structural coupling model and outputs the optimization results of the printing structure parameters and the initial path configuration;

[0027] A control strategy training module that constructs a reinforcement learning environment based on the optimization results and trains a control strategy for dynamically regulating the printing process;

[0028] A path and support generation module that generates a layer-by-layer printing path and a support structure configuration according to the optimization results and the control strategy;

[0029] A printing execution and closed-loop control module that executes the 3D printing process based on the support structure configuration and the layer-by-layer printing path, collects printing status data, and performs real-time adjustment according to the control strategy.

[0030] The present invention provides an environmentally friendly packaging box rapid prototyping method and system based on biodegradable materials.

[0031] Having the following beneficial effects:

[0032] 1. By constructing a thermal-structural coupling finite element model based on biodegradable materials, the present invention can accurately simulate the changes in the temperature field, stress field, and thermal strain field during the printing process. Therefore, it can predict the performance of the material under heat conduction and mechanical loading at the design stage, thereby optimizing the printing path and the support structure configuration, significantly improving the printing accuracy, avoiding structural instability caused by thermal deformation and interlayer stress, and ensuring the overall strength and stability of the printed object.

[0033] 2. By adopting a multi-objective optimization algorithm and comprehensively considering factors such as printing time, thermal field uniformity, structural stiffness, and material usage, the present invention realizes the collaborative optimization of the printing path and the structural design. On the premise of ensuring the printing quality and structural strength, by precisely controlling the material usage and printing time, the utilization rate of resources and the efficiency of the printing process are maximized, material waste is reduced, and it conforms to the concept of environmental protection design.

[0034] 3. By introducing a reinforcement learning control strategy, the present invention can collect data in real time during the printing process and perform dynamic adjustment. Specifically, based on state variables such as the real-time temperature of the print head, the material extrusion rate, and the printing speed, the system can adjust the printing parameters in real time, thus being able to cope with complex changes in the printing environment and timely adjust key parameters in the printing process, such as the heating power and the printing speed, reducing the risk of interlayer cracking and warping, and ensuring that each layer of printing is accurate and error-free.

[0035] 4. By combining structural topology optimization and thermal simulation feedback, the present invention has made innovations in the support structure configuration. The layout of the support structure gives priority to thermally deformed sensitive areas and overhanging structure areas, and can be dynamically adjusted according to real-time data, which not only improves the printing stability, but also effectively reduces the waste of support materials, reduces the energy consumption and material consumption during the printing process, and further enhances the overall economy and sustainability of printing. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a schematic flow chart of the method of the present invention;

[0037] Figure 2 is a schematic diagram of the system architecture of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the specification of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0039] To better understand the present invention, the above content will be described in detail below with specific embodiments.

[0040] Please refer to the attached Figure 1 , an embodiment of the present invention provides an environmentally friendly packaging box rapid prototyping method based on biodegradable materials, including the following steps:

[0041] Collect the thermal conductivity, specific heat capacity, elastic modulus, and density parameters of the selected biodegradable material, and establish a thermal-structural coupling finite element model based on this material;

[0042] In this embodiment, aiming at the problems of complex thermal behavior and easy deformation of biodegradable materials during the rapid prototyping process of environmentally friendly packaging boxes, thermal physical property parameter modeling is first carried out based on the selected biodegradable materials.

[0043] Specifically, the thermophysical parameters include, but are not limited to, thermal conductivity, specific heat capacity, elastic modulus, and density. Among them, thermal conductivity and specific heat capacity are key parameters in heat transfer calculations, used to describe the heat transfer ability of materials per unit time and the ability of unit mass to increase temperature after absorbing heat. The elastic modulus is used to characterize the deformation ability of materials under stress, while density is used for heat capacity calculation and analysis of mass-related quantities. The above parameters can all be obtained through material databases, literature, or experimental measurements.

[0044] It should be noted that due to the drastic temperature changes during the printing process of biodegradable materials, their thermophysical parameters often exhibit significant temperature dependence. Therefore, in this embodiment, when establishing the thermo-structural coupled finite element model, a temperature-dependent material constitutive relationship is introduced. For example, parameters such as thermal conductivity, specific heat capacity, and elastic modulus are all constructed as function models that vary with temperature. This functional relationship can be constructed through polynomial fitting, experimental interpolation, or look-up table methods, and has a certain degree of selectivity.

[0045] The thermo-structural coupled finite element model constructed in this embodiment integrates the steady-state and transient heat conduction equations, the structural mechanics response model, and the material thermal strain model. Specifically, in the thermal analysis module, the following heat conduction control equation is used for temperature field simulation:

[0046]

[0047] Where:

[0048] is the temperature variable;

[0049] ρ(T) is the density of the selected biodegradable material at temperature T;

[0050] c(T) is the specific heat capacity of the selected biodegradable material at temperature T;

[0051] k(T) is the thermal conductivity of the selected biodegradable material at temperature T;

[0052] Q is the internal heat source term;

[0053] The above equation can be solved by finite element discretization to obtain the transient temperature field distribution within the entire printing domain.

[0054] In the part of structural mechanics, to achieve coupling with the thermal field, a structural statics model dominated by thermal strain is adopted. Its basic control equation is:

[0055]

[0056] Where, is the stress tensor, f is the body force term, and the stress-strain relationship is established based on the thermoelastic constitutive model:

[0057] σ = D(T):[ε - ε th (T)];

[0058] where D(T) is the stiffness matrix related to temperature, ε is the total strain, and ε th (T) is the thermal strain, which is defined as:

[0059] ε th (T) = α(T)·(T - T0)·I;

[0060] where α(T) is the thermal expansion coefficient of the selected biodegradable material at temperature T, T0 is the reference temperature, and I is the unit tensor.

[0061] In the actual calculation process, thermal analysis and structural analysis are synchronously calculated through time steps to form a two-way thermal-structural coupling. In the specific implementation, the current temperature field is first solved within each time step, and based on this temperature field, the elastic modulus and thermal strain are updated. Then, the current structural response is solved by inputting into the structural module, and finally, the corresponding stress field and thermal strain field are output.

[0062] Through the simulation solution of this model, the temperature field evolution, internal thermal stress distribution, and thermal deformation trend induced by temperature difference during the 3D printing process of the structure to be printed can be obtained. The output temperature field, stress field, and thermal strain field provide an input basis for subsequent path optimization and support structure generation, and also provide a state variable simulation environment for the training of reinforcement learning control strategies.

[0063] And the finite element model can optionally be implemented in commercial simulation platforms (such as ANSYS, COMSOL), or discrete solutions can be implemented based on a self-developed code framework. The model discretization method can be selected as linear or nonlinear elements, and the mesh can be locally refined for key regions to improve the simulation accuracy.

[0064] Therefore, the thermal-structural coupling finite element model in this embodiment not only fully considers the non-linear change characteristics of material properties under temperature changes, but also provides computable physical field outputs, laying a solid foundation for subsequent structural optimization and control strategy construction.

[0065] Based on the thermal-structural coupling model, a multi-objective optimization model with printing time, thermal field uniformity, forming strength, and material consumption as objectives is constructed to optimize the parameters of the structure to be printed packaging, and the structural design parameters and initial printing path configuration are obtained;

[0066] In this embodiment, after the construction of the thermal-structural coupling finite element model of the biodegradable material is completed, a multi-objective optimization modeling for the target structure is further carried out based on this model to realize the comprehensive coordinated design of the 3D printing parameters of the environmentally friendly packaging box.

[0067] The multi-objective optimization model comprehensively considers four factors: printing efficiency, temperature control stability, structural performance, and material consumption. The set optimization objectives include: minimizing printing time, maximizing thermal field uniformity, maximizing the stiffness of the formed structure, and minimizing material usage. The above optimization objectives are used to characterize the printing requirements of speed, stability, and low resource consumption, and to seek the optimal solution combination within the printing parameter space.

[0068] Among them, the printing time is evaluated by dividing the total length of the printing path by the printing speed per unit time, and is usually related to the printing layer thickness and filling rate.

[0069] The thermal field uniformity is quantified by the maximum temperature difference between layers in the model prediction results, which is derived from the transient temperature field output by the thermal-structural coupling finite element model.

[0070] The structural stiffness is measured by the effective modal stiffness or equivalent elastic modulus of the target structure under given boundary conditions, and comes from the output of the structural field.

[0071] The material usage is calculated by integrating the volume of the filled area and the volume of the support structure, combined with the density, and is an important indicator for controlling the printing resource cost.

[0072] There are potential conflict relationships among the above multiple optimization objectives. For example, increasing the layer thickness may shorten the printing time, but at the same time it will affect the heat conduction efficiency and printing strength. To achieve an overall consideration of each objective, in this embodiment, an optional multi-objective optimization algorithm is selected for model solution, such as the non-dominated sorting genetic algorithm (NSGA-II) or the multi-objective particle swarm optimization algorithm (MOPSO), which can be flexibly selected according to the implementation platform and the scale of the example.

[0073] In the optimization model, the design variables are the key structural parameters that make up the printed structure, mainly including:

[0074] The printing layer thickness, as the printing step control parameter along the Z-axis direction, affects the printing rate and the structural connection strength.

[0075] The filling density, which represents the proportion of material in the non-solid area, has a significant impact on weight and material consumption.

[0076] The local wall thickness variation can be used to enhance the rigidity of the key stress-bearing areas and suppress thermal deformation.

[0077] The ventilation structure layout parameters, which involve the position and size of ventilation holes or heat conduction auxiliary structures, have a direct impact on the heat conduction path and the heat release rate.

[0078] In addition, for a specific packaging structure, other structural parameters can also be optionally introduced as design variables, such as the layout form of the support structure (including support type, spacing between support contact points, etc.) and the overall geometric profile characteristics (such as fillet radius, connection slope, etc.). The ranges and initial values of the above design variables can be set according to the structure type, printing material, and printer parameters.

[0079] The optimization problem can be formally expressed as follows:

[0080] Four optimization objective functions are constructed as follows:

[0081] F(x) = [f1(x), f2(x), f3(x), f4(x)];

[0082] Among them, the F(x) is the objective-vector, which contains multiple sub-objective functions, namely f1(x), f2(x), f3(x), and f4(x).

[0083] Specifically, it includes:

[0084] Objective 1: Minimize the printing time:

[0085]

[0086] Among them, f1(x) is the printing time objective function, which is used to measure the total time required to complete all printing paths under the current design variable x; x is the design variable, representing the set of printing parameters to be optimized, including but not limited to: printing layer thickness, filling density, local wall thickness, and ventilation structure layout parameters, etc.; L path (x) is the total length of the printing path corresponding to the design parameter x, v(x) is the instantaneous printing speed function when the print head moves on the path, and s is the path length integral variable, representing the path coordinates from the starting point 0 to the end L path of the printing path.

[0087] Objective 2: Minimize the temperature difference between layers:

[0088]

[0089] Among them, f2(x) is the thermal field uniformity objective function, which represents the maximum temperature difference between layers during the printing process of the structure. This objective function reflects the stability of the heat conduction process during printing and needs to be minimized; T i (x) and T j (x) respectively represent the instantaneous temperatures of the i-th layer and the j-th layer under the current design variable x, and this temperature is obtained from the results of thermo-structural coupling simulation. The calculation points can be at a certain critical moment during the printing process; i and j represent the structure layer numbers.

[0090] Objective 3: Maximize the structural stiffness:

[0091]

[0092] Among them, f3(x) is the structural stiffness objective function, representing the average load-bearing capacity of the overall structure, and V is the maximization objective; the three-dimensional volume region of the printed structure, corresponding to the geometric boundary range of the structure; σ eq (x) is the equivalent stress distribution function, often calculated using the von-Mises equivalent strength criterion, reflecting the stress response of the structure under the given load; dV is the spatial volume element, used to integrate the stress results at all points in the structural domain V.

[0093] Objective 4: Minimization of material usage:

[0094] f4(x) = ∫Vρ(x)dV;

[0095] Among them, f4(x) is the material usage objective function, representing the total material mass or volume consumed for printing the current structure, which is the objective function to be minimized; ρ(x) is the local material density function, usually depending on geometric features such as filling density, whether it is support, cavity, solid, etc. in the structure, and can be a constant or a function related to position and design parameters; V is the same as f3(x), and dV represents the structural solid volume region; it represents the volume element in the integral, and the sum after accumulation gives the total material distribution within the entire structure.

[0096] Therefore, by constructing a thermal-structural coupling analysis model, the mutual influence relationship between the temperature field and the structural response during the printing process is systematically characterized. Furthermore, taking the printing time, thermal field uniformity, structural stiffness, and material usage as optimization objectives, a multi-objective optimization model is established to jointly solve the design parameters and path configuration of the packaging structure to be printed. This model not only fully considers the coupling balance among printing efficiency, heat conduction stability, forming strength, and resource consumption, but also realizes the parameter collaborative optimization for printing performance, structural function, and environmental constraints by introducing key variables such as layer thickness, filling density, wall thickness variation, and ventilation structure layout into the optimization dimension. Finally, the model can output a set of structural design solutions and initial printing path configuration schemes that meet the comprehensive performance requirements, providing a clear parameter basis and model support for subsequent path refinement, printing control strategy formulation, and structural function realization.

[0097] Based on the structural design parameters, an enhanced learning environment is constructed. The state variables during the printing process are used as inputs, and the control variables are used as actions. The printing control strategy is trained using the policy optimization algorithm to achieve the dynamic regulation of the printing process;

[0098] In this embodiment, based on the solution results of the thermal-structural coupling finite element model and the multi-objective optimization model of the aforementioned biodegradable material, an intelligent dynamic control system for 3D printing of packaging structures was further constructed to achieve fine control and real-time optimization of the printing process.

[0099] The dynamic control system uses reinforcement learning (RL) as the core intelligent decision-making method. By simulating the layer-by-layer printing behavior of biodegradable materials, it optimizes the adjustment strategy of key parameters during the printing process. The construction of the reinforcement learning environment is based on the results of multi-physics field coupling simulation, systematically describing the transient evolution characteristics of the temperature field and stress field during the printing process, and ensuring the physical consistency and prediction accuracy of the model.

[0100] In this embodiment, the state space design of the reinforcement learning environment fully considers multi-source state information during the printing process, including but not limited to the following five main state variables:

[0101] Real-time temperature of the printing nozzle:

[0102] It reflects the instantaneous temperature under the current working state of the nozzle, and is affected by the coupled influence of the nozzle heating power and the printing speed. The dynamic change of temperature directly affects the fluidity of the material and the layer bonding quality;

[0103] Material extrusion rate:

[0104] It describes the linear velocity of the printing material extruded from the nozzle. This rate is related to the printing path, nozzle diameter, and material viscosity, and is a key indicator for controlling the printing thickness and density;

[0105] Printing speed:

[0106] It represents the linear velocity of the print head moving along the path. The adjustment of the printing speed directly affects the uniformity of material spreading and the heat conduction characteristics;

[0107] Hot bed temperature:

[0108] As the printing substrate, the temperature control of the hot bed directly affects the material cooling rate and the bottom layer adhesion effect, preventing warping and interlayer peeling;

[0109] Printing layer height:

[0110] It refers to the Z-axis step height during single-layer printing, which is directly related to path planning and filling density. An overly large or small layer height will affect the printing accuracy and material stacking.

[0111] The above state variables are used as the observation inputs of the reinforcement learning model. Through the real-time monitoring and update of the thermal-structural coupling model, they accurately reflect the evolution of the material's heat absorption, flow, curing, and mechanical characteristics during the printing process.

[0112] In the reinforcement learning strategy, based on the observation results of the above state variables, a corresponding control action space is defined, including:

[0113] Nozzle heating power:

[0114] Control the output power of the heating element inside the nozzle, which determines the melting state of the material and the extrusion flow rate;

[0115] Air cooling speed:

[0116] Adjust the cooling rate of the printing area through an external fan, which affects the curing time of the material and the interlayer bonding effect;

[0117] Print head speed:

[0118] Dynamically adjust the moving rate of the print head along the path to achieve precise control of the printing accuracy and the thermal accumulation effect;

[0119] Feeding rate:

[0120] Adjust the supply speed of the material from the feeding end to the nozzle, and synchronously adjust it with the extrusion of the nozzle to ensure the uniformity of the printed layer.

[0121] During the optimization process of the reinforcement learning strategy, the above control variables are used as adjustable actions and are called by the policy network in real time. Through the adaptive exploration and learning of the reinforcement learning algorithm, various parameters are continuously adjusted to optimize the printing quality and efficiency.

[0122] Specifically, the training of the reinforcement learning control strategy uses the Proximal-Policy-Optimization (PPO) algorithm. Its main optimization goal is to maximize the cumulative reward function J(θ), that is, to maximize the long-term reward expectation value of the agent in a given state:

[0123]

[0124] where θ is the parameter of the policy network;

[0125] R t is the immediate reward at time step t, defined as the feedback obtained from the transition from the current state to the next state;

[0126] γ is the discount factor, representing the decay weight of future rewards;

[0127] T is the maximum number of steps in a training episode.

[0128] The PPO algorithm adjusts the policy network to maximize the cumulative reward. Specifically, in each training iteration, the agent selects actions according to the current policy and obtains rewards based on the environmental feedback. Based on this reward information, the policy update process is as follows:

[0129]

[0130] Among them, θ t represents the policy parameter at time step t;

[0131] θ t+1 represents the policy parameter at time step t + 1;

[0132] α is the learning rate, which is used to control the step size of policy parameter update;

[0133] is the gradient of the policy loss function L(θ) with respect to the policy parameter θ.

[0134] And the loss function L(θ) for policy optimization is defined as:

[0135]

[0136] Among them, r t (θ) is the policy update rate; is the advantage function, which is used to measure the superiority of the current action and reflects the difference between the reward in the state and the average expected reward; ∈ is the clipping parameter, which is used to limit the change range of the policy.

[0137] Advantage function is composed of the value function and the reward difference, and is estimated by the following formula:

[0138]

[0139] Among them, is under the policy the state value function of state s t represents the expected future cumulative reward obtained by executing the policy starting from this state;

[0140] Therefore, through continuous iterative calculation and update, the PPO algorithm effectively balances exploration and exploitation in each round of training, enabling the policy to achieve dynamic optimization and adaptive adjustment of parameters during the printing process.

[0141] And through multiple rounds of simulation iteration, the reinforcement learning agent continuously selects the optimal action according to the current state (i.e., optimizes control parameters such as the nozzle heating power, air cooling speed, printing speed, and feeding rate), and optimizes and updates the policy according to the feedback reward. This process is strengthened and verified with the support of the thermal-structural coupling simulator, thus ensuring the efficiency and stability in a complex printing environment.

[0142] Finally, through sufficient strategy convergence and simulation verification, this embodiment successfully implements a set of efficient, stable, and adaptive 3D printing control strategies. This strategy can dynamically adjust multiple key parameters during the printing process, thereby optimizing the printing quality, improving material utilization, and effectively reducing unstable factors during the printing process.

[0143] Generate a layer-by-layer printing path and a support structure configuration plan that match the printed object according to the structural design parameters, printing path configuration, and printing control strategy;

[0144] In this embodiment, based on the comprehensive calculation results of the thermal-structural coupling finite element model, multi-objective parameter optimization model, and reinforcement learning strategy optimization model of the aforementioned biodegradable material, a path generation module and a support structure automatic configuration module for 3D printing task execution are further constructed to generate a layer-by-layer printing path and a support structure configuration plan that match the structural form and process characteristics of the printed object.

[0145] Specifically, to ensure the rationality of the printing path and the stability of the overall structural performance, the path generation process first performs preliminary layer division based on the structural design parameters and the 3D geometric model. Specifically, a slicing algorithm is used to equally space-slice the 3D model along the building direction (usually the Z-axis direction), and the cross-sectional contour of each layer is extracted as the basic data for path planning. This cross-sectional data is preprocessed through path adjustment and continuity reconstruction algorithms to remove path discontinuities, sharp corners, and invalid closed segments, thereby ensuring path smoothness and improving printing stability.

[0146] During the path generation process, considering the coordination of the overall structural performance and local process responses, the result of structural topology optimization is introduced as the basis for path correction in this embodiment. Structural topology optimization is a method of reconstructing material distribution based on objective functions and constraint functions, which can identify the main load-bearing paths and unnecessary structural regions with stress concentration under load conditions. Combining this optimization result, the path planning module can locally densify or sparsely adjust the printing path to make the path distribution more in line with the mechanical response requirements, especially to achieve a more reasonable material stacking plan in the stress-concentrated areas.

[0147] Furthermore, to eliminate the thermal stress concentration and structural deformation caused by the interlayer temperature gradient, the path generation also needs to comprehensively consider the thermal simulation feedback information. In this embodiment, based on the thermal-structural coupling simulation model, a multi-time-step transient analysis is performed on the layer-by-layer printing process, and the high-gradient temperature regions and potential deformation regions that appear during the heat conduction, heat convection, and material curing processes of each layer are extracted. High-gradient regions usually have a significant thermal shrinkage trend, which is likely to cause warping or internal stress residue of the printed structure.

[0148] Therefore, in this embodiment, the heat simulation feedback results are introduced into the path layout strategy for guiding optimization. Specifically, in the heat deformation sensitive area, the path generation strategy preferentially introduces methods such as reducing the path spacing and changing the stacking path direction to form a temperature release zone and reduce the heat accumulation rate. And in the arrangement of path turning and starting points, combined with the thermal and cold boundary conditions, an asymmetric path sorting strategy is implemented to weaken the stress concentration caused by heat convection.

[0149] In terms of the configuration of the support structure, in this embodiment, by integrating the geometric features of the printed structure, the results of topology optimization and the heat simulation feedback results, the spatial distribution and contact mode of the support structure are intelligently inferred and adjusted. First, in the area with overhanging structures (i.e., the overhanging angle exceeds the critical value of material self-support), the system automatically generates linear or surface support members, and the support members are connected to the main structure by a controllable gap or anchoring method for later removal or structural linkage.

[0150] Secondly, in the areas highly sensitive to thermal deformation, such as local sharp corners, protrusions, thin-walled sections, etc., due to insufficient heat convection conduction, thermal warping deformation is likely to occur. To suppress such problems, in this embodiment, the layout of the support structure gives priority to forming a constraint system in such areas, specifically including applying structural support to the main direction of thermal stress, arranging heat dissipation channels or forced cooling auxiliary components at key positions to guide the balanced distribution of the heat flow path.

[0151] Furthermore, the configuration of the support structure does not adopt a static template, but is iteratively generated layer by layer according to the dynamic generation results of the printing path. After each layer of path is generated, it can be judged whether new, continuous or terminated support members are needed according to the topological changes and heat conduction boundary changes between this layer and the next layer. The "adaptive evolution" of the support structure is realized through the way of layer-by-layer feedback, improving the overall matching and printing stability.

[0152] In the above process, the linkage configuration of the printing path and the support structure is completed by the parallel calculation of the path planning module and the support configuration module, and is coordinated and synchronized by the printing control strategy optimization module. The reinforcement learning control strategy adopted in this embodiment can provide a predictive input based on the historical thermal stress change trend in the path generation stage, which is used to evaluate the influence of different path layouts on the subsequent temperature distribution, and then used as an auxiliary control reference for path planning.

[0153] For example, when the control strategy recognizes that a specific path direction will cause heat to concentrate at the boundary, the printing order or segmentation strategy of this path can be adjusted in advance to avoid the problem of thermal stress accumulation from the source. The support configuration module can also adjust the activation time and density of the support based on this thermal evolution trend, so as to realize the dynamic scheduling and optimization of the support use.

[0154] In summary, the layer-by-layer paths generated in this embodiment not only match the structural design parameters and the geometric features of the printed object, but also have the ability to respond and adjust to the changes in thermal stress distribution and structural stability. On the basis of meeting the printing support function, the support structure pays more attention to the matching with the thermally deformed sensitive areas and the integration of thermal feedback, so as to achieve a high degree of adaptability and process stability in the overall printing process.

[0155] Execute the 3D printing process according to the layer-by-layer printing path and the support structure configuration scheme. During the process, collect the printing status data in real time and perform closed-loop adjustment according to the control strategy.

[0156] In this embodiment, during the 3D printing process, the printing status data is collected in real time and closed-loop adjustment is performed according to the control strategy. This closed-loop adjustment process mainly includes real-time monitoring and feedback control. By dynamically adjusting parameters such as temperature, pressure, extrusion flow rate, and interlayer misalignment, the stability of the printing process and the final printing quality are ensured.

[0157] Real-time data collection

[0158] During the printing process, the system collects key printing status data in real time through multiple sensors, including thermocouple sensors, infrared thermometers, and displacement sensors. Specifically:

[0159] The thermocouple sensor is used to detect the temperature of the print head and the printing bed to ensure the temperature stability during each layer printing process.

[0160] The infrared thermometer is used to detect the temperature distribution between each layer, especially in areas with large heat accumulation.

[0161] The displacement sensor monitors the misalignment between layers during the printing process and detects possible interlayer misalignment or structural instability.

[0162] Closed-loop control and adjustment formula

[0163] After obtaining these data, perform closed-loop adjustment on the printing process based on the control strategy to reduce possible printing defects, especially interlayer cracking and warping. To achieve this goal, the system adjusts the key parameters in the printing process according to the following formula:

[0164] Temperature adjustment formula:

[0165] For the temperature of each printing layer, the control system controls the heat flow by adjusting the printer heating parameters. The temperature adjustment formula can be expressed as:

[0166] ΔT l =α(T set -T l );

[0167] Among them, ΔT l is the temperature adjustment value, T set is the target temperature, and T l is the temperature of the current printing layer. α is the temperature control gain coefficient, which controls the heat input. By adjusting α, the system can control the temperature of the print head or the bed to reduce the thermal stress problem caused by excessive or too low temperature differences.

[0168] Interlayer misalignment adjustment formula:

[0169] The displacement sensor monitors the interlayer misalignment during the printing process in real time. Assuming that during the printing process of a certain layer i, the interlayer misalignment is δ i , then the printing speed or extrusion flow rate is adjusted to reduce the misalignment. The interlayer misalignment adjustment formula is:

[0170]

[0171] Among them, δ i is the misalignment value of the current printing layer, and δ i+1 is the expected misalignment value of the next layer. is the extrusion flow rate, Δt is the printing time step, and β is the control coefficient, which reduces the misalignment value by adjusting the flow rate in real time.

[0172] Extrusion flow rate adjustment formula:

[0173] The extrusion flow rate directly affects the deposition speed and uniformity of the printing material. When interlayer misalignment occurs, the control system adjusts the extrusion flow rate to correct the material deposition speed. The flow rate adjustment formula is as follows:

[0174]

[0175] Among them, is the adjusted extrusion flow rate, δ i is the actual interlayer misalignment of the current printing layer, and δ target is the target misalignment value. γ is the adjustment coefficient, which makes the deposition of each layer more uniform and reduces the occurrence of misalignment by optimizing the adjustment range of the flow rate.

[0176] In the actual printing process, based on the above formulas, the system continuously performs closed-loop adjustment according to the real-time collected data. For example, when the infrared thermometer detects that the temperature of a certain layer is too low, the system will increase the heating power of the print head; if the displacement sensor monitors that the interlayer misalignment is too large, the extrusion flow rate or printing speed will be adjusted to reduce the misalignment problem. At the same time, the adjustment of the printing parameters is based on the control gain coefficients (such as α, β, γ), and these coefficients are dynamically adjusted according to the feedback data during the printing process.

[0177] Through real-time feedback control, the system can continuously adjust the key parameters during the printing process, ensuring that each layer of the printing process is completed as expected, thereby reducing the occurrence of interlayer cracking, warping, and other printing defects. Especially in complex structures or high-precision printing tasks, the closed-loop adjustment scheme can significantly improve the printing quality and avoid structural instability caused by inaccurate temperature or flow control.

[0178] Therefore, in this embodiment, by introducing a closed-loop adjustment mechanism, combining real-time data acquisition and intelligent control strategies, the key parameters are dynamically adjusted during the 3D printing process to optimize the printing process, improve the printing accuracy, and effectively reduce problems such as interlayer cracking and warping.

[0179] Please refer to the attached Figure 2 , in this embodiment, an environmentally friendly packaging box rapid prototyping system based on biodegradable materials is also provided, including:

[0180] A material modeling module that establishes a thermal-structural coupling finite element model based on the thermal conductivity, specific heat capacity, elastic modulus, and density parameters of biodegradable materials;

[0181] A structural optimization module that constructs a multi-objective optimization model based on the thermal-structural coupling model and outputs the optimization results of the printing structure parameters and the initial path configuration;

[0182] A control strategy training module that constructs a reinforcement learning environment based on the optimization results and trains a control strategy for dynamically regulating the printing process;

[0183] A path and support generation module that generates a layer-by-layer printing path and a support structure configuration according to the optimization results and the control strategy;

[0184] A printing execution and closed-loop control module that executes the 3D printing process based on the support structure configuration and the layer-by-layer printing path, collects printing status data, and performs real-time adjustment according to the control strategy.

[0185] In this embodiment, an environmentally friendly packaging box rapid prototyping system based on biodegradable materials is provided. By comprehensively utilizing advanced thermal-structural coupling finite element models, multi-objective optimization algorithms, reinforcement learning control strategies, and real-time closed-loop control, the system can effectively improve the accuracy and efficiency of the 3D printing process. The system mainly includes the following modules:

[0186] Material modeling module: This module establishes a thermal-structural coupling finite element model based on the basic physical properties of biodegradable materials such as thermal conductivity, specific heat capacity, elastic modulus, and density. This model can accurately simulate the heat conduction and mechanical behavior of the material during the printing process, predict the deformation and stress distribution of the material under different temperature fields, and provide accurate data support for subsequent optimization design.

[0187] Structure Optimization Module: This module further constructs a multi-objective optimization model through a thermal-structural coupling model, aiming to optimize the design parameters of the printed structure. Multi-objective optimization not only considers performance indicators such as structural strength and stiffness but also comprehensively takes into account factors such as thermal stress and thermal shrinkage, thereby generating a printed structure design with excellent performance and outputting the optimization results of the initial path configuration. These results provide a basis for subsequent path planning and the design of support structures.

[0188] Control Strategy Training Module: Based on the optimization results, the system constructs a reinforcement learning environment. By training the control strategy, it can dynamically adjust the printing parameters during the printing process. The reinforcement learning model continuously optimizes in this environment, learns through the feedback of historical printing states, and gradually improves its adaptability to complex changes during the printing process to ensure printing quality and process stability.

[0189] Path and Support Generation Module: This module generates the layer-by-layer printing path and support structure configuration according to the optimization results and control strategy. The design of the printing path not only considers the rationality of material deposition but also adjusts according to the mechanical requirements and thermal stress distribution of the structure to ensure the stability and adhesion of each layer of printing. The support structure is precisely arranged in the overhanging areas and thermally deformed sensitive areas that may occur during printing according to the model optimization results to provide necessary support and avoid warping or structural instability during the printing process.

[0190] Printing Execution and Closed-loop Control Module: During the printing process, the system performs 3D printing according to the generated support structure configuration and layer-by-layer printing path. During the printing process, it real-time collects printing state data, including key parameters such as temperature, displacement, and extrusion flow rate, and makes real-time adjustments according to the preset control strategy. The core function of this module is to dynamically adjust the printing parameters according to the real-time data feedback during the printing process, such as adjusting the extrusion flow rate, printing speed, etc., to ensure that each layer during the printing process reaches the best state and reduce risks such as interlayer misalignment and thermal deformation.

[0191] Through the collaborative work of the above modules, the entire system realizes the full-process optimization and control from material modeling, structure optimization to printing execution, and can significantly improve the printing accuracy, stability, and efficiency of the environmentally friendly packaging box, meeting the requirements of modern environmental protection design and rapid production.

[0192] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A rapid prototyping method for an environmentally friendly packaging box based on biodegradable materials, characterized in that, It includes the following steps: Collect the thermal conductivity, specific heat capacity, elastic modulus and density parameters of the selected biodegradable material, and establish a thermal-structural coupling finite element model based on this material; Based on the thermal-structural coupling model, construct a multi-objective optimization model with printing time, thermal field uniformity, forming strength and material usage as the objectives, optimize the packaging structure parameters to be printed, and obtain the structural design parameters and the initial printing path configuration; Based on the structural design parameters, construct a reinforcement learning environment, use the state variables during the printing process as inputs and the control variables as actions, and train the printing control strategy using a policy optimization algorithm to achieve dynamic regulation of the printing process; According to the structural design parameters, printing path configuration and printing control strategy, generate a layer-by-layer printing path and a support structure configuration plan that matches the printing object; Execute the 3D printing process according to the layer-by-layer printing path and the support structure configuration plan, collect the printing state data in real time during the process, and perform closed-loop adjustment according to the control strategy.

2. The rapid prototyping method of an environmentally friendly packaging box based on biodegradable materials according to claim 1, characterized in that, The thermal-structural coupling finite element model considers the thermal conductivity, specific heat capacity and elastic modulus of the biodegradable material that change with temperature during the printing process, and outputs the temperature field, stress field and thermal strain field.

3. A rapid prototyping method for an environmentally friendly packaging box based on biodegradable materials according to claim 1, characterized in that, The multi-objective optimization model is solved using a multi-objective optimization algorithm, and the optimization objectives include: Minimizing the printing time; Minimizing the temperature difference between layers; Maximizing the structural stiffness; Minimizing the material usage.

4. A rapid prototyping method for an environmentally friendly packaging box based on biodegradable materials according to claim 3, characterized in that, The design variables of the optimization model include the printing layer thickness, filling density, local wall thickness change and ventilation structure layout parameters, and the packaging structure parameters include the printing thickness, support structure layout, wall thickness and ventilation structure configuration.

5. A rapid prototyping method for an environmentally friendly packaging box based on biodegradable materials according to claim 1, characterized in that, The state variables include the real-time temperature of the printing nozzle, the material extrusion rate, the printing speed, the hot bed temperature and the printing layer height; the control variables include the nozzle heating power, the air cooling speed, the printing head speed and the feeding rate.

6. A rapid prototyping method for an environmentally friendly packaging box based on biodegradable materials according to claim 5, characterized in that, The reinforcement learning control strategy is trained using the proximal policy optimization algorithm, and the policy convergence is achieved through iterative simulation of the printing process.

7. A rapid prototyping method for an environmentally friendly packaging box based on biodegradable materials according to claim 4, characterized in that, The printing path is generated based on the results of structural topology optimization and thermal simulation feedback, and the support structure is preferentially arranged in the thermally deformed sensitive area and the overhanging structure area.

8. A rapid prototyping method for an environmentally friendly packaging box based on biodegradable materials according to claim 1, characterized in that, The closed-loop adjustment is carried out by collecting the printing temperature, pressure, extrusion flow rate and interlayer misalignment in real time, and adjusting the printing parameters based on the printing control strategy to reduce the risk of interlayer cracking and warping.

9. A rapid prototyping method for an environmentally friendly packaging box based on biodegradable materials according to claim 8, characterized in that, The collection of the printing state data includes thermocouple sensors, infrared thermometers, and displacement sensors.

10. An environmentally friendly packaging box rapid prototyping system based on biodegradable materials, according to any one of claims 1-9, a rapid prototyping method for an environmentally friendly packaging box based on biodegradable materials, characterized in that, It includes: A material modeling module that establishes a thermal-structural coupling finite element model based on the thermal conductivity, specific heat capacity, elastic modulus and density parameters of the biodegradable material; A structural optimization module that constructs a multi-objective optimization model based on the thermal-structural coupling model and outputs the optimization results of the printing structure parameters and the initial path configuration; A control strategy training module that constructs a reinforcement learning environment based on the optimization results and trains a control strategy for dynamically regulating the printing process; A path and support generation module that generates a layer-by-layer printing path and a support structure configuration according to the optimization results and the control strategy; The printing execution and closed-loop control module executes the 3D printing process based on the support structure configuration and the layer-by-layer printing path, collects the printing status data, and performs real-time adjustment according to the control strategy.

Images